A DECADE OF SYSTEMIC TREATMENT FOR HEPATOCELLULAR CARCINOMA: EXPERIENCE OF A TERTIARY CENTER

Introduction: A decade ago, sorafenib became the first agent to achieve a survival improvement in patients (pts) with hepatocellular carcinoma (HCC) and remains the cornerstone of systemic therapy. Recently, novel drugs have shown positive results, although they are still not widely available. Clinical trials strictly include pts with preserved liver function and good performance status (PS), what not always mirror the profile from reallife practice. In view of the potential incorporation of novel agents, data from clinical practice may support therapeutic decisions. We aimed to describe a decade of HCC treatment, focusing on clinical features, outcomes and prognostic factors. Methods: A database of HCC pts from Instituto do Câncer do Estado de São Paulo was analyzed. Descriptive statistic was used for categoric variables; overall survival (OS) was estimated by Kaplan Meier and compared by log rank. Cox regression was used to define prognostic factors. Results: From 2008 to 2018, 239 pts started sorafenib. Median age was 61.7 years (18.9-83.1), 112 (48.1%) had hepatitis C, 45 (19.3%) had hepatitis B and 36 (15.5%) had alcoholic hepatopathy. The majority presented BCLC-C (n = 197, 83.7%) and Child-Pugh (CP)A (n = 205, 86.5%). Median treatment duration was 4.9 months (1.1-23.8) and median OS was 9.8 months (95%CI 7.9-11.6). Considering only the subgroup with CPA and ECOG-PS 0/1 (n = 201), the median OS was 10.4 months (95%CI 8.5-12.5). CPA vs B (p = 0.006); neutrophil-to-lymphocyte ratio < 3.5 vs ≥ 3.5 (p < 0.0001); decrease of ≥ 25% in alpha-fetoprotein vs no decrease (p = 0.004) and disease control vs progression (p = 0.006) were independently associated with better OS. At sorafenib discontinuation, 48.9% of the patients were CPA and 51.1% had ECOG-PS 0-1. Second-line therapies were provided for 52 (21.7%) patients. Median post-progression OS (from sorafenib discontinuation) was 4.5 months (95%CI 3.5-5.4). CP B-C vs A (p = 0.004), PS 3-4 vs 0-2 (p = 0.04) and discontinuation due to progression vs toxicity (p < 0.001) were associated with worse post-progression OS. Conclusion: Our cohort presented similar OS to that reported in phase III studies. At progression, half of the patients would not meet inclusion criteria for trials, mainly due to clinical and liver function deterioration. Radiologic response, alpha-fetoprotein, hepatic dysfunction and biomarkers of inflammation can help to predict outcomes and select pts that could potentially benefit from second-line therapies.